FIGURE CAPTIONS
Figure 1. Location of proposed drill areas (open boxes) and previous DSDP and ODP sites (solid dots).Figure 2. A. Preferred viscosity structure used to calculate hotspot motion from Steinberger and O'Connell (1998). A low-viscosity upper mantle is used to reproduce the Hawaiian-Emperor bend. A high-viscosity lower mantle is employed; otherwise the relative motions between hotspots are greater than observations. The Harvard tomographic model S12WM13 was used to infer mantle density heterogeneities. The gradual increase in viscosity was chosen to minimize disagreements with models based on postglacial rebound, which mainly constrain viscosity in the upper half of the mantle. B. The predicted motion of the Hawaiian plume between 90 and 43 Ma after Steinberger (2000). The model predicts a southward component of motion ~10 mm/yr. This results from the mantle flow at depth, which also tends to have a southward component of the same magnitude partly due to a return flow opposite to Pacific plate motion assumed in the model. The model predicts only a small relative motion between the Hawaiian and Louisville hotspots, in accordance with the age progressions observed along the two hotspot tracks. Other models with a lower viscosity in the lower mantle predict substantially higher flow speeds and substantially larger southward motion of the Hawaiian hotspot.
Figure 3. Evidence for compaction-induced inclination error in Pacific deep-sea sediments from Tarduno (1990). A. DSDP sites with sediment-based paleomagnetic data obtained using thorough demagnetization techniques. B. Test for bias in sediment-based paleomagnetic inclinations using 27 age groups. Io is the observed inclination.Ie is the expected inclination derived from nonsediment sources. A least-squares fit yields a slope, f = 0.52. A delete-1 jackknife resampling shows that the data reject the hypothesis of zero flattening (f = 1) at the 95% confidence interval. C. Inclination error vs. expected inclination for f = 0.5 (circles). Open squares = the maximum error in inclination caused by a 5° error in the reference pole, where p is the colatitude. Filled squares = the combined effect of inclination shallowing caused by compaction and an error in the reference pole.
Figure 4. Inclination errors caused by sediment compaction plotted vs. expected inclination values. Curves show the relationship tan Io = f tan Ie, where Io = the observed (measured) paleomagnetic inclination, Ie = the expected inclination and f = a variable describing the degree of compaction-induced inclination shallowing. A value of f = 0.52 was derived from paleomagnetic analyses of deep-sea sediments from the Pacific plate (Tarduno, 1990). A. The gray box shows the range of inclination values expected for the fixed-hotspot vs. moving-hotspot hypotheses. The expected inclination value for a fixed hotspot is derived from Hawaii's current position, whereas values for Nintoku, Detroit, and Meiji are based on the hypothesis that their location along the Emperor trend records mainly motion of the Hawaiian hotspot. Because the hotspot-motion hypothesis predicts these Emperor seamounts formed at mid latitudes, errors in sedimentary inclinations induced by compaction will be near their maximum potential values, assuming flattening factors similar to those derived from Cretaceous deep sea sediments from the Pacific. B. The expected difference in inclination between the fixed-hotspot and moving-hotspot models (gray box and horizontal dashed lines) shown against compaction-induced inclination error curves. Given a flattening factor of 0.52, the potential error in sediment-based inclination is larger than the signal of hotspot motion proposed for testing for Detroit and Meiji seamounts, the error is two-thirds of the signal proposed for testing.
Figure 5. A. Average inclination values for three inclination-group models from Detroit Seamount. Errors are 95% confidence interval. Also shown is the predicted inclination at 81 Ma based on prior Pacific apparent polar wander path (APNP) poles (Gordon, 1983). B. Paleolatitude values with 95% confidence intervals for the inclination groups. Also shown is the present-day latitude of the Hawaiian hotspot (black line). C. Estimated angular dispersion (S) of the inclination groups (black line) shown vs. the predicted values for 45-80 Ma (dark field) and 80-110 Ma (light field) from McFadden et al., (1991). D. Orthographic projection of the colatitude (primary) for Detroit seamount (star). The colatitude is distinct at the 99% confidence level (shaded) from previous 81-82 Ma poles (ellipses). Poles are derived from the following: 81 Ma (Gordon, 1983); 82 Ma (Sager and Pringle, 1988); 33n (79.1-73.6 Ma) (Vasas et al., 1994). The sense of offset between the NRM data and the demagnetized (primary) data is the same as that between the new paleolatitude result and results based on prior Pacific pole positions. This is the effect expected if these previous pole positions are contaminated by secondary magnetizations. Figure is after Tarduno and Cottrell (1997). VGP = virtual geomagnetic pole.
Figure 6. Plot of latitudinal distance from the 43-Ma bend in the Hawaiian-Emperor hotspot track vs. age (light circles). Age data are not available for Meiji, Tenchi, and Jimmu; their positions, based on a constant latitudinal progression, are shown for reference. Dark gray circles indicate positions after the difference between the present-day latitude of the 43-Ma bend and Hawaii is subtracted from each of the present-day latitudes of the Emperor seamounts. In effect, we slide the Emperor trend down the Hawaiian chain so that the bend coincides with the position of Hawaii (inset). This reconstruction allows the following test. If the Emperor seamounts record mainly motion of the Hawaiian hotspot, paleolatitudes should fall close to this corrected latitudinal trend; if the hotspot has been stationary, the paleolatitudes should fall close to the present-day latitude of Hawaii. Triangles indicate the paleolatitudes of Suiko and Detroit Seamounts, with their 95% confidence intervals. The null hypothesis that the paleolatitude result from Suiko is drawn from the same population as the Detroit data can be rejected at the 95% confidence level using nonparametric tests (Kolmogorov - Smirnov). In the absence of a rotation of the entire solid Earth with respect to the spin axis, known as true polar wander (Tarduno and Cottrell, 1997; Cottrell and Tarduno, 2000b; Tarduno and Smirnov, 2001), the hotspot may have moved continuously southward at a rate of 30 50 mm/yr while the plate also drifted slowly northward (dark gray). Figure is after Tarduno and Cottrell (1997).
Figure 7. A. Estimates of zonal quadrupole Gauss coefficient ( g20) relative to the axial dipole ( g10) from Livermore et al. (1984). Pacific data are rotated using a fixed hotspot reference frame (see model "B" in Livermore et al., 1984). Our proposed sampling covers the range where Livermore et al. propose a change in sign of the quadrupole term. B. Paleointensity determined from studies of submarine basaltic glass (SBG) compiled by Juarez et al. (1998). The proposed sampling covers the transition from the Cretaceous Normal Polarity Superchron (K-N) to the Late Cretaceous Cenozoic mixed polarity interval. VADM = virtual axial dipole moment.
Figure 8. Compositional changes in magmas produced by the Hawaiian hotspot through time. The shaded field shows the range of published 87Sr/86Sr of tholeiitic basalts vs. age and distance along the Hawaiian-Emperor chain. Note that data from Detroit seamount are significantly less radiogenic than at younger volcanoes. The crossed circles connected by the thick dotted line shows the trend in age difference between seamounts and the underlying ocean crust (from Keller et al., 2000).
Figure 9. Proposed site locations (note change in scale): (A) Meiji Guyot, (B) Detroit Seamount, (C) Nintoku Seamount, (D) Ojin Seamount, and (E) Koko Guyot.